Semantic Generalization of the Conditioned Eyelid Response with Two Types of Training

Abstract

Tests of semantic generalization to the word VASE, and phonetic generalization to the word EARN, visually presented, were carried out subsequent to discrimination Training with URN reinforced and TOP not reinforced in one group, and following straight 100% reinforcement of URN in another group (40 Ss each), using standard eyelid conditioning procedures. Generalization of extinction to the words URN and TOP was finally tested in both groups. On the semantic generalization test, half of the Ss received VASE as the first test stimulus and half received EARN. There was no evidence for semantic generalization as opposed to phonetic generalization following discrimination training, and only slight evidence for such generalization following straight conditioning that appeared on the first semantic generalization test trial only. Order of test stimuli produced more significant effects than did their nature. On the generalization of extinction tests to the words URN and TOP, the discrimination training group showed no evidence that the semantic generalization testing reduced the response level—in short, no generalization of extinction as well as no semantic generalization. The straight conditioning groups, however, did show some diminution of responding or generalization of extinction from the semantic generalization testing, and this was statistically significant.